Air Force Institute of Technology
AFIT Scholar
Theses and Dissertations Student Graduate Works
3-24-2016
Neck Injury Criteria Development for Use in
System Level Ejection Testing; Characterization of
ATD to Human Response Correlation under -Gx
Accelerative Input
Craig M. Zinck
Follow this and additional works at:https://scholar.afit.edu/etd Part of theOccupational Health and Industrial Hygiene Commons
This Thesis is brought to you for free and open access by the Student Graduate Works at AFIT Scholar. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of AFIT Scholar. For more information, please [email protected].
Recommended Citation
Zinck, Craig M., "Neck Injury Criteria Development for Use in System Level Ejection Testing; Characterization of ATD to Human Response Correlation under -Gx Accelerative Input" (2016). Theses and Dissertations. 420.
NECK INJURY CRITERIA DEVELOPMENT FOR USE IN SYSTEM LEVEL EJECTION TESTING; CHARACTERIZATION OF ATD TO HUMAN
RESPONSE CORRELATION UNDER -GX ACCELERATIVE INPUT
THESIS
Craig M. Zinck, Captain, USAF AFIT-ENV-MS-16-M-194
DEPARTMENT OF THE AIR FORCE AIR UNIVERSITY
AIR FORCE INSTITUTE OF TECHNOLOGY
Wright-Patterson Air Force Base, Ohio DISTRIBUTION STATEMENT A.
The views expressed in this thesis are those of the author and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government.
AFIT-ENV-MS-16-M-194
NECK INJURY CRITERIA DEVELOPMENT FOR USE IN SYSTEM LEVEL EJECTION TESTING; CHARACTERIZATION OF ATD TO HUMAN RESPONSE
CORRELATION UNDER -GX ACCELERATIVE INPUT THESIS
Presented to the Faculty
Department of Systems Engineering and Management Graduate School of Engineering and Management
Air Force Institute of Technology Air University
Air Education and Training Command In Partial Fulfillment of the Requirements for the Degree of Master of Science in Systems Engineering
Craig M. Zinck, MBA Captain, USAF
March 2016
1 AFIT-ENV-MS-16-M-194
NECK INJURY CRITERIA DEVELOPMENT FOR USE IN SYSTEM LEVEL EJECTION TESTING; CHARACTERIZATION OF ATD TO HUMAN
RESPONSE CORRELATION UNDER -GX ACCELERATIVE INPUT
Craig M. Zinck, MBA Captain, USAF
Committee Membership: Lt Col Jeffrey C. Parr, PhD
Chair
Michael E. Miller, PhD Member
Chris Perry, MS Member
2 AFIT-ENV-MS-16-M-194
Abstract
The use of Helmet Mounted Displays is becoming ubiquitous in the field of aviation, adding operational capability while increasing head-supported weight and potential neck injury risk during ejection. Developing neck injury criteria to evaluate and quantify neck injury risk is important to ensure ejection systems are produced within acceptable safety standards. In this study, an ATD to human transfer function for is developed that quantifies the difference between Anthropomorphic Test Device (ATD) and human neck response data from -Gx accelerative tests, and demonstrates how this transfer function can be applied to ATD test data to make previously developed human risk functions directly applicable to interpreting the ATD data with a human-based neck injury criterion. To gain an understanding of how the MANIC(Gx) can be applied to escape system testing, the ATD test values were evaluated using the current state of the art MANIC(Gx) human risk curves. A difference between the human and ATD MANIC(Gx) neck response was measurable, the ATD indicated lower MANIC(Gx) levels at
equivalent AIS risk probabilities. For instance, at 5% probability of neck injury risk, the ATD MANIC(Gx) (for AIS 2+ and AIS 3+ probability of injury) values were 0.29 and 0.364 respectively where the human values at the same injury percentage was 0.56 and 0.72. The associated injury criteria can be directly applied to ATD safety testing of aircraft ejection or vehicles systems in -Gx accelerative loading to directly translate ATD neck load results to the probability of human injury.
3
Acknowledgments
I would like to express immense gratitude to all of my fellow students for their
encouragement and assistance during my time at AFIT, I would not have made it without many of you. I would also like to thank my committee members and my advisor for their unwavering guidance throughout this program. Mostly, I would like to thank my
amazing wife and kids for the sacrifices that they have made over the past 18 months in order for me to succeed in this endeavor; I love you all so very much.
4 Table of Contents Page Abstract ...2 Acknowledgments...3 Table of Contents ...4 Table of Figures ...6 List of Tables ...7 I. Introduction ...8 Background...8 Problem Statement ...14 Justification...15 Thesis Outline ...17 Assumptions/Limitations ...18 Expected Contributions ...19
II. Literature Review ...20
Introduction ...20
Head and Neck Anatomy and Biomechanics ...21
Injury Classification ...25
Neck Injury Criteria ...28
Statistical Methods ...39
5
III. Comparison of Human and ATD Neck Response to Frontal Impact (-Gx)
Acceleration ...41 Chapter Overview ...41 Abstract...41 Introduction ...42 Background...45 Methods ...49
Results and Discussion ...52
References ...57
IV. Development of a Frontal Impact (-Gx) Neck Injury Criterion ...58
Chapter Overview ...58 Introduction ...60 Background...63 Methods ...70 Results ...77 Discussion...83
V. Conclusions and Recommendations ...86
Chapter Overview ...86
Summary of Research ...86
Investigative Questions Answered ...91
Significance of Research ...92
6
Final Thoughts ...93
Bibliography ...95
Table of Figures Figure 1. F-35 HMD Ejection Risks vs. Body Weight ... 17
Figure 2. Human Neck Anatomy (Grierson and Dunn) ... 23
Figure 3. Neck Loading Resultant Movements (Grierson and Dunn) ... 24
Figure 4. Probability of AIS 2 or Greater NHTSA and Human Neck Injury Risk Curves (Parr et al., 2013) ... 38
Figure 5. Probability of AIS 3 or Greater NHTSA and Human Neck Injury Risk (Parr et al., 2013) ... 38
Figure 6. Human AIS 2+ and AIS 3+ Risk Curves (Parr et al., 2013) ... 39
Figure 7. Linear Regression of ATD and Human MANIC(Gx) Data ... 53
Figure 8. Human Linear Regression Data Plot with Regression through the Origin ... 55
Figure 9. ATD Linear Regression Data Plot ... 55
Figure 10. ATD and Human MANIC(Gx) Linear Regression Model of Expected Values Across Accelerative Inputs ... 79
7
List of Tables
Table 1. MANIC Components (Gx, Gy, Gz) and limits (Parr, 2014) ... 35
Table 2. MANIC Risk Function Input Matrix (Parr, 2014) ... 37
Table 3. MANIC(GX) ATD Critical values ... 51
Table 4. ATD and Human MANIC(Gx) Linear Regression Model Equations ... 53
Table 5. MANIC(Gx) Critical Values ... 66
Table 6. ATD Test Parameters ... 74
Table 7. Human AIS 2+ comparison of human and ATD-transformed MANIC(Gx) values and associated injury risk levels ... 78
Table 8. Human AIS 3+ Comparison of human and ATD-transformed MANIC(Gx) values and associated injury risk levels ... 78
Table 9. MANIC(Gx) Predicted Values at 5%, 10% and 22% Probability of Injury for Human, ATD-adjusted, and NHTSA risk curves ... 82
8
Neck Injury Criteria Development for Use in System Level Ejection Testing; Characterization of ATD to Human Response Correlation Under -Gx Accelerative
Input I. Introduction Background
Manned aircraft of the future will certainly utilize Helmet Mounted Device (HMD) technologies in an effort to ensure they are state-of-the-art; maintaining their operational high-ground. HMDs provide a pilot with an interface that requires less visual scanning, resulting in reduced object capture time, and increased user efficiency.
Incorporation of HMDs replaces legacy displays with in-visor instrument displays along with other operationally relevant information such as radio frequencies, weapon data, threat data, and flight safety warnings. A focus on Human Systems Integration is imperative when designing systems that are intended to work in-concert with the user (Parr 2014; Rash et al. 2009). Supporting additional head-borne mass may be an effective use of pilot resources, but also may negatively affect performance if the weight change is enough to hinder operational ability. The additional weight of HMDs must be evaluated so that the potential benefits and injury risks can be quantified in order for the decision maker to have clarity of action when deciding on HMD alternatives. Designing escape systems that opt to use HMD hardware to increase performance while adding significant head-supported weight has presented the U.S. Air Force (USAF) with a unique personnel safety issue.
9
Escape systems incorporated into the design of fighter aircraft are integral to ensuring pilot safety. Safety is of utmost importance to the USAF, the escape system developers, and the pilots who operate these weapon systems. Neck injury evaluation related to the function of escape systems, particularly with added HMD weight, has been a topic for many years. In 1995 a phase one Small Business Innovative Research (SBIR) defined the need for a human injury prediction tool that could be used to correlate
manikin to human neck response when under accelerative load (Grierson and Dunn, 1995); therefore, a significant amount of HMD development effort should be dedicated to acquiring safe escape system equipment. This research intends to provide decision makers with a method that enables developmental testing in the -Gx plane of accelerative input to be accomplished using Anthropometric Test Devices (ATD) but with a human-based neck injury criterion supported by a quantitative risk functions.
Limiting injuries by creating relevant design requirements is the goal of the USAF, however, a logical methodology for creating HMD requirements focused on pilot safety has proven difficult to quantify. There has been incongruence between acceptable and unacceptable testing conditions when using existent criteria. A more quantitative method of escape system evaluation is required for better informed decision making that is founded on quantified human injury risk. There have been designs that fail to meet all requirements, yet when a SME qualitative analysis deems that the limit excursions are negligible or “within tolerance” they are ushered through. Current helmet device design limitations consist of weight and center of gravity (CG) restrictions imposed by an
10
interim criterion called the Knox Box, developed 21 years ago, long before the
advancement of current HMD designs. However, development of the Knox Box assumed a single ejection seat with a known range of accelerations and a specific range of
airspeeds, which affect forces due to windblast. The complex relationship between helmet, ejection seat, and airspeed parameters makes the development of easily-measured helmet requirements difficult to specify without overly restricting the design trade space.
Rather than specifying specific helmet characteristics, system level requirements can be developed and enforced against the prime contractor to permit trades in helmet characteristics, ejection system characteristics and ejection procedures. These system level requirements can provide system requirements to be demonstrated during system verification and validation. Multiple neck injury criteria have been used in aviation, focusing on a wide range of accelerative input conditions that mimic specific portions of forces absorbed by a pilot during ejection. Focus areas of these previous studies include characterizing head and neck biomechanics, tensile and bending neck strength
characterization, injury classification, lower neck, upper neck, and finally multi-axial neck criteria development (Eppinger et al. 1999; Parr et al. 2013; Parr, 2014; FAA, 2011; Parr et al., 2015)
Numerous neck injury criteria will be briefly explained in the following literature review, to establish the history of neck injury study and its application to aviation. The Knox Box, developed by USAF ejection researchers (Perry, Buhrman, & Knox III, 1993)
11
who were developing HMD prototype designs. The Knox Box permits helmets having a range of mass and CG values as its primary metrics, creating a virtual box that all HMD prototype designs should fall within. The idea behind the Knox Box was to limit the weight and off-axis moment (CG) to minimize chances of pilot injury. This criterion was developed as an early design guideline, not intended for long-term use in HMD
development efforts however, it is still being utilized as the starting point for HMD design (Parr et al., 2014).
The US Navy and Air Force developed a 12-part Neck Injury Criterion (NIC) that was developed by using instrumented aerospace ATDs of varying sizes configured with Hybrid III necks. Those ATDs were then subjected to sled test ejections at different equivalent airspeeds (Nichols, 2006). The NIC provides criteria that can indicate limit overages however, after SME review these excursions are sometimes waived and the test is given a pass. Essentially, no definite or quantitative pass fail criteria are adhered to within the NIC, and none of the twelve elements of the NIC are supported by adequate human neck injury risk functions (Parr et al., 2014).
All of the aforementioned aviation studies have moved the USAF research and operational testing communities toward finding an escape system evaluation tool, however, there are still areas that need improvement. In 2009 the Air Force Life Cycle Management Center (AFLCMC) escape system oversight office requested the
development of an improved criterion for evaluating pilot neck injury risk using a multi-axial evaluation tool that accounts for occupant size and is based upon human neck injury
12
data. Parr implemented a previously proposed pilot-scale neck injury criterion called the multi-axial neck injury criterion (MANIC), that is relevant to and developed for
evaluating neck injury risk in the aviation environment. However, further investigation in each axis (Gx, Gy, and Gz) is required to verify the use of the MANIC in developmental testing with ATDs. This research will investigate how the human and ATD neck
response to acceleration level relate to one another using -Gx input forces similar to frontal impact (coronal plane force).
During initial prototyping, an HMD system is unable to be thoroughly tested in an operationally representative environment (Parr, 2014). The field of aviation neck injury biomechanics seeks to provide more information about weight and center of gravity restrictions, how these design considerations affect the human neck, and how to incorporate these considerations into future helmet design prior to prototyping. The current method used in the USAF to assess the design of prototype helmets is the Knox Box, a method that was intended as a temporary tool. Even though it is considered a temporary tool, the Knox box provides boundaries for design teams early in the process. If these bounds were imposed on helmet design teams after prototyping has begun, the cost repercussions would be much more severe. Only after the HMD is technologically mature can it be tested on ATDs using sled test ejections at the escape system level (including ejection seat and pilot ensemble). The cost of system level testing can be prohibitive; this creates a need for a well-defined neck injury criterion that can be used to alleviate redesign later in a design programs life cycle. Having the ability to use the
13
results from a system level ATD ejection sled test for human injury prediction would be valuable to all stakeholders involved.
Parr (2014) attempted to account for the discrepancies in the above methods and fulfill the updated neck injury requirements set forth by AFLCMC. Parr’s research resulted in the development of the MANIC, which sought to include all six major forces and moments that could be observed in the upper neck as a result of accelerative loading in the three primary axes of acceleration (Gx, Gy, and Gz). The MANIC complies with AFLCMCs multi-axial neck injury criteria requirement and is defined in Equation 1. Multi-Axial Neck Injury Criteria (Parr, 2014)
Equation 1. Multi-Axial Neck Injury Criteria (Parr, 2014)
The MANIC is a robust evaluation tool that is directly linked to associated human injury through generation and analysis of human risk functions . This criterion fulfills all of the requirements imposed by the AFLCMC; no other neck injury metric/criterion developed to date does this.
This research intends to provide decision makers with a method for using ATD test results as an indicator of human injury during developmental testing when exposed to similar -Gx accelerative input. The proposed method will be accomplished using existing human-based MANIC(Gx) quantitative risk functions.
14
Problem Statement
Historically, the use of ATDs with instrumented necks, Post Mortem Human Subjects (PMHS), and porcine subjects have been accelerated to levels that would be injurious to humans in order to quantify the injury possibilities associated with individual test setup parameters (Cheng et al., 1986; Bass et al., 2006; Salzar et al. 2009; Beeman et al. 2013; Eppinger et al., 1999). Human testing has limitations, accelerating human subjects to a potential condition of known injury is unethical. Impact testing with human subjects has acceleration magnitude limits (for example, 10G in the z-axis) in order to keep potential injury of test volunteers minimized. The creation of a human to ATD transfer function and the development of new comparison methods for human and ATD -Gx accelerative loading results to quantify neck injury risk to aircrew is the intent of this research. Providing information that affords systems engineers the foundation to define and generate valid, traceable requirements to future HMD designs based on ATD data will save time, coordination and cost. Leveraging past research efforts is essential for success in this endeavor. By using the current aviation specific MANIC criteria and its associated MANIC(Gx) human risk curves, this research will concentrate on the frontal plane (coronal) forces for development of a -Gx ATD to human transfer function. The following research questions will be explored in this thesis effort:
1) What is the difference in expected MANIC(Gx) between human/PMHS and ATDs over the range of -Gx accelerative input observed from previous laboratory
15
2) Can the observed differences in peak MANIC(Gx) be used to create a transfer
function to make the Parr et al. human-based risk functions and associated neck injury criterion more appropriate for use in testing with ATDs?
Defining injury parameters that are directly linked to validated human risk curves will provide HMD developers a metric to refine their designs.
Justification
In 2000 the Chief of Staff of the Air Force (CSAF) initiated acceptance of a congressional mandate that was previously approved to allowed females to fly fighter aircraft by expansion of ejection seat weight ranges resulting in lowering the low limit from 132 lbs. to 103 lbs. and expanding the upper limit from 211 lbs. to 245 lbs.
(Personal Communication, 2014). Incorporation of additional head supported weight in the form of HMD technology and expanded weight ranges ensure that future weapons systems have the best available equipment and best available aircrew in future weapon platforms. However, a current weapon platform undergoing the acquisition process, the F-35, has elected to make a safety decision and limit pilot weights to above 136 lbs., eliminating the pilot pool below this limit due to increased injury potential.
Aircraft escape systems are integral to the safety of aircrew. Any redesign in these systems could drive schedule slip, aircraft availability issues and create an operational disadvantage. If these re-designs are due to escape system safety concerns, a more exhaustive engineering effort to fix the error will increase costs and leave a temporary gap in operational capability. Systems engineering principles have shown that early
16
funding decisions are the most critical, any changes (re-design) late in the program can cause severe budget overrun (US DoD, 2013).
The added weight of HMDs is also a concern, particularly for the smaller aviator. Generation II HMDs add 2.19 lbs. of head supported weight to the standard helmet, and the F-35 Generation III designs add 2.60 lbs. These HMD weight increases make up a larger proportion of head supported weight for pilots at the small end of the
anthropometric range (Perry, 1999). By widening the acceptable weight range, the USAF accepted a higher risk for aircrew below 136 lbs., by proxy. According to current USAF guidance; future escape systems and any associated equipment should not increase the current risk of pilot injury.. The associated risks, by body weight, for USAF F-35 ejections are depicted in Figure 1, this graphic depicts hazard risk index (HRI) as a function of aircrew weight and ejection air speed measured in knots equivalent airspeed (KEAS). These accepted risks directly affect over 33 percent of the female pilot
population (Meyer & Andries, 2014).
The creation of a human to ATD transfer function will be used for the development of a method to link ATD test results to human injury likelihood using previously developed human neck injury risk functions at specific injury levels that are applicable to all weight ranges. This will enable decision makers to justify concrete escape system requirements by providing them with quantitative injury risk data in the prototyping and developmental testing process, allowing all stakeholders involved to accurately create, test, and certify HMDs and escape systems.
17
Figure 1. F-35 HMD Ejection Risks vs. Body Weight
Thesis Outline
This thesis is structured in the scholarly format. Chapter 1 provides an overview of the research topic, research questions, and motivation. Chapter 2 provides a literature review of the pertinent scholarly research in the field of neck injury criteria and injury biomechanics. Chapter 3 is a conference paper that has been accepted for presentation at the 2016 Industrial and Systems Engineering Research Conference (ISERC). This
ISERC paper documents the first phase of the research conducted in this thesis effort to develop a transfer function based on existing data from previous human and ATD -Gx
110 130 150 170 190 210 230 250 0 100 200 300 400 500 600 Aircrew Nude Weight (lbs.) Airspeed (KEAS) F‐35 Escape System Envelope W/ HMD >4.8 Lbs. Low Risk High Risk Ejection Forces W/ Generation II helmet High risk Airspeed
18
accelerative tests. Chapter 4 is a journal article to be submitted to the Journal of
Aerospace Medicine and Human Performance (called the Journal of Aviation, Space, and Environmental Medicine prior to January 2015). It documents the complete research effort undertaken in this thesis, starting with the -Gx human to ATD transfer function development, followed by the application of this human to ATD transfer function to calculate new ATD-transferred human risk probabilities using the previously developed human MANIC(Gx) risk functions that allow for evaluation of escape system neck injury risk due to -Gx acceleration directly with ATDs. Chapter 5 provides a summary and conclusions of the thesis research.
Assumptions/Limitations
Although the AFLCMC escape system oversight office requirement is to create neck injury criteria that are multi-axial, a human-based, ATD-transformed neck injury criterion in a single plane of motion will be used as a proof of concept. Future research will concentrate on the other critical planes of accelerative input (Gy and Gz). A transfer function to predict likelihood of ejection neck injury is based on the correlation of ATD neck response to that of human subjects. In past research it has been shown that ATDs neck response gets progressively more flexible when accelerative loads are applied when compared to PMHS response at high G levels (Beeman et al., 2013). The assumption that human muscle responds relatively similar to this stiffness under loading is required to ensure that data can be compared between the two populations across a wide range of -Gx accelerative input (up to 45 G’s). Parr’s pilot-scale, human-based MANIC risk
19
functions and associated neck injury criteria had to make the assumption that human neck response was approximately equal to ATD neck response to apply the MANIC injury risk limits to testing with ATDs (Parr, 2014). This work seeks to investigate Parr’s
assumption and attempt to make the -Gx portion of Parr’s MANIC more applicable to testing with ATDs. Future work will require additional ATD testing to validate the applicability of the transfer functions created here for use in the systems level ejection environment. Additionally, any helmet design requirements or restrictions that can be gleaned from this data should conform to the Air Force requirement generation protocols listed in the Defense Acquisition Guidebook (DAG). Finally, the evaluation of chronic injury due to neck loading caused by HMD use will not be covered in this research.
Expected Contributions
The transfer functions and MANIC(Gx) risk functions used in this work will be applicable for use in military safety testing applications that experience x-axis
accelerations and may be applicable to the automotive or other industries where head supported weight is a requirement.
20
II. Literature Review Introduction
Biomechanics of the neck when subjected to accelerative loading has been studied by multiple organizations, including The National Highway Traffic Safety
Administration (NHTSA), the Federal Aviation Administration, academic research institutions and the Department of Defense (DoD) (Buhrman and Perry, 1994; Bass et al., 2006; Eppinger et al., 1999; Parr et al., 2013; Parr 2014;Parr, 2015; Salzar et al., 2009). The automotive industry (NHTSA) research efforts were some of the first to attempt to characterize the effects of accelerative loading on human neck by testing surrogate subjects. The testing of PMHS, porcine subjects, and ATDs all played a part of the initial testing, evaluation, neck injury classification, and injury criterion creation. This chapter will serve as a background for the research to be conducted in this thesis as well outline other contributions that are relevant to this research.
The DoD adopted some of the practices that were set forth in the automotive realm and adapted those practices in an attempt to characterize human neck response during ejection. Multiple neck injury criteria have since been developed to quantify neck loading and will be addressed in this chapter. It has proven difficult to extend the
automotive centered analysis to human risk during ejection. Among the difficulties is the fact that the NHTSA has been most concerned about accelerations resulting from frontal impact crashes and therefore has concentrated their efforts to quantify the risk to vehicle occupants due to forces in the Gx plane, while pilots can experience substantial force in combinations of all three axes during the ejection sequence. Injury probability and
21
classification of injury severity during ejection are the stakeholder level metrics used to quantify the risks future pilots might incur during ejection from an aircraft and are therefore important to project during acceptance testing of any DoD aircraft. These anticipated risks must be scalable and reliable across a wide range of pilot anthropometry as well as statistically sound, reliable, and quantitative in nature. Injury criteria and quantification protocols that are useable at escape system level accelerations are the anticipated end goal of DoD neck injury biomechanics research. Engineering
decomposition and analysis of the problem in each force vector may lead to development of functional pieces of a complete human based escape system evaluation tool that can be aggregated, or used as stand-alone vector specific criteria.
Risk curve development is aimed at specifying a likelihood of injury based on input factors (neck size, HMD weight, HMD CG, injury classification) and loading of a pilot’s neck. A test manager or decision authority will be able to use the developed risk curves and make a critical safety decision early in development based on quantitative analysis, ideally prior to incurring the substantial costs of live ejection testing in the developmental and validation testing phases.
Head and Neck Anatomy and Biomechanics
“The neck and spinal cord system could be the most complicated human physiology aspect to evaluate during crash impact” (Mertz, 1971)
The ability to use an Anthropomorphic Test Device (ATD) in crash tests allows the test agency to subject the ATD to forces that could not be applied to humans due to
22
the risk of human injury. For this reason, military test agencies have used the ATD to test and qualify escape systems since the inception of the ejection seat. ATD capabilities have been improved since the first ATD testing took place, and new neck components have been developed to mimic the movements and forces present in the human neck under loading. A more human representative neck called the Hybrid III was originally developed by General Motors and adopted for use in frontal impact studies by the NHTSA (Rasmussen & Plaga, 1993). Some have postulated that the Hybrid III neck under loading is much stiffer than a PMHS neck under similar loads and therefore is not a biofidelic representation of human neck response (Myers et al.,1991, Bass et al., 2006). Due to the observed differences in human and PMHS versus ATD neck response, this research attempts to construct a model that links ATD and human neck response. ATDs have been the best surrogate available for mimicking human response in operational testing at higher accelerative levels. A test agency should be able to use an ATD and predict how the resultant ATD data would affect a human in similar conditions.
Unfortunately, this has proven difficult in the extremely dynamic aviation environment with ejection seat equipped aircraft.
Upper neck loads are the focus in the aviation domain since the lower neck has more muscle mass surrounding the area and the vertebrae are relatively larger. Previous testing has indicated upper neck injury when PMHS are subjected to frontal impact (Mertz et al., 1978). The upper neck critical values are lower and more injuries have been observed in this area than in the lower neck as addressed in previous human and PMHS testing (Nichols, 2006). Figure 2 depicts the cervical section of the spine
(C1-23
C7). The upper neck area is defined by the location where the Occipital Condyle (OC), the base of the skull, and the atlas (C1) vertebrae connect (Grierson and Dunn, 1995). This connection provides the ability to articulate the head forward and backward at the base of the skull (Flexion and Extension). A vertebra called the axis (C2) connects below the atlas via a bony projection called the dens (Odontoid Process) and provides the ability of left and right rotation. The remaining Neck vertebrae provide the neck the ability to move in flexion, extension and, lateral bending with minor rotational allowances.
Figure 2. Human Neck Anatomy (Grierson and Dunn)
The majority of neck injury takes place in the upper neck when subject to combined axis loading (Cheng et al., 1982). Figure 3 displays bending, compression, tension and shear forces in the neck. NHSTA studies that led to development of
24
automotive neck injury criteria (The Nij; explained in a later section) deemed that the most significant neck injury mechanism in frontal impact (-Gx accelerative input) are axial loading (tension or compression) and front/back bending moment (flexion or extension).
Figure 3. Neck Loading Resultant Movements (Grierson and Dunn)
Previous research indicates that a smaller neck is at a load bearing disadvantage compared to a larger neck due to the musculoskeletal composition of the neck and the scaling associated with neck circumference and vertebrae size (Salzar et al., 2009). Therefore, a larger neck is expected to be more effective at handling larger loads than smaller necks. Injuries however, have been shown to vary between test subjects due to
25
anthropometric factors like body weight, neck size and neck strength. Biofidelity
agreement is further complicated when attempting to obtain human neck injury tolerance information. Indirect methods such as testing with human volunteers at low accelerations, cadaver testing at higher accelerations, computer simulation, and crash testing may be used to provide data that is similar to human response. However, this data may not be representative of human neck response when exposed to large accelerative input and should not indicate that ATDs can be a direct replacement in all ejection test scenario (Eppinger et al., 1999). Due to the previously explained restrictions that are inherent to testing with humans and the costs associated with PMHS testing; the Hybrid-III ATD with an instrumented neck has been made available in multiple sizes to account for anthropometric differences. Even though these steps have been taken to account for a majority of anthropometric differences, there have been studies that show the hybrid-III neck is not representative of the human neck, and some that show PMHS data could be skewed for the same reasons (Herbst et al., 1998). PMHS testing has also shown that incongruences exist between cadaver and human neck response. However, at present an ATD fitted with a Hybrid-III neck seems to be the best surrogate that can be used to evaluate neck response under accelerative loading, particularly during ejection system and helmet qualification testing.
Injury Classification
In 2008, the Association for Advancement of Automotive Medicine (AAM) met to finalize and classify injuries into a common scale that resulted in the creation of the Abbreviated Injury Scale (AIS). The AIS is an anatomically based, consensus derived,
26
global severity scoring system that classifies each injury by body region according to its relative importance on a 6-point ordinal scale with the least sever injury labeled AIS 1 and the most severe labeled as an AIS 6 (AAM, 2008). In general AIS 1 is a minor injury, AIS 2 is a moderate injury, AIS 3 is a serious injury, AIS 4 is a severe injury, AIS 5 is a critical injury, and AIS 6 is a maximal injury (AAM, 2008). The AIS provides a means for scoping analysis effort as well as a concise way to present injury data to decision makers once it has been analyzed. Using a standard injury scale, gives the escape system and HMD developers, testers, and stakeholders a standardized injury definition to work from. The AIS also quantifies and provides a common understanding of injury level and how the injuries would affect the aircraft or vehicle operator. It has been proven useful to motor vehicle crash investigators when attempting to identify a mechanism of injury to aid in vehicle design. The USAF has applied the AIS in a similar manner as the motor vehicle industry. This research will utilize the AIS injury severity assessment scales.
The AIS classification is used to assess neck injury based on a universal scale. Quantifying the injuries by specific attributes allows a researcher to evaluate the forces that may lead to that specific level of injury. This classification is used as a guide for neck injury and is an integral part of risk function creation. The Air Force Life Cycle Management Center (AFLCMC) escape system oversight office has set the limit for future USAF escape systems at a 5% chance of AIS 2 or greater neck injury (White, J. E. Personal communication; May 2012).
27
In the field of injury biomechanics, difficulty lies in assessing AIS injury classifications when applying ATDs. Where injury is readily evident in human and PMHS testing via analysis after the accident or accelerative experiment, ATD “injury” is an elusive result. The advantage of ATD testing is evident when attempting to measure neck loads and when testing above safe levels for humans (Rash et al., 1998). One can only refer to the measured loads and the anticipated maximum allowable human loads to infer an injurious test condition.
AIS classifications are universally used for many industries, including the automotive, aviation, and medical fields. AIS categories are used to describe and assess injuries through a common medical terminology across the range of disciplines. For instance, PMHS are subjected to AIS injury inspection after being exposed to high levels of accelerative input to ascertain and codify any observed neck injuries.
The AIS levels that will be used as evaluation scenarios in this research are the AIS 2 and AIS 3 levels. AIS 2 injuries have a 1-2% probability of death associated with this classification injury and are coded as moderate, AIS 3 injuries are described as serious and have an 8-10 % probability of death associated with them (AAM, 2008). Based on retroactive analysis of ejection data in the USAF, 69% of injuries have been associated with injuries higher than AIS 3 (Grierson and Dunn, 1995). The following sections will highlight various neck injury criteria that are germane to ATD and human -Gx acceleration evaluation.
28
Neck Injury Criteria Nij
Neck injury in the aviation community has become an area of interest since the incorporation of HMD technology. The advancement of HMDs has forced stakeholders to study the modes of injury in all 12 degrees of freedom. Formulated in the automotive industry, the Nij, can be related to aviation modes of injury. NHTSA has been studying the injurious conditions that are present in frontal vehicular impacts for decades and have a vast amount of data and test results available for review by the aviation industry. Ejection has distinct phases; catapult, windblast, wind drag/drogue chute deceleration, parachute opening shock and free fall. Much research has focused on the catapult phase where predominant resulting forces are compressive in nature. Not as much attention has been given to the other input forces experiences during ejection. Assuming the ejection seat does not rotate during ejection, wind drag and drogue chute ejection forces are equitable to frontal impact forces where flexion motions are created through a -Gx input. Tensile forces are presented in a less intuitive manner, as the body slows, the cranial inertia pulls it further away from the body, exacerbated by more extreme flexion.
The development of neck injury criteria within the automotive industry was initiated by the proliferation and ubiquitous incorporation of restraint systems and
emphasis on crash testing (FMVSS No. 208). The automotive sector experienced a rise in death associated with operating vehicles that led to government regulations demanding that vehicles become safer. The incorporation of airbag technology led to the need for performance limits created for mid-sized male, small-sized occupants, and children,
29
leading to emphasis on testing and injury prevention during frontal (-Gx) impact (Eppinger et al., 1999). It has been shown that there are similar modes of neck injury prevalent in automotive accidents and aircraft ejection (Nichols, 2007, Salzar et al., 2009). Similar to the automotive industry, the DoD fighter aircraft communities are interested in supporting aviator health through the design of safer escape systems capable of accommodating smaller, and larger individuals than are currently supported (Nichols, 2006). Reducing injury to the aircrew, viewed as the most important part of the weapon system, has surfaced as the primary issue to AF safety decision makers.
Neck injury criteria are defined for a number of different conditions such as injury mechanism, acceleration environment, and impact condition (Bass et al., 2006). The Mertz criteria were used in the automotive environment and are scalable to account for neck size (Mertz, 1993; Mertz et al., 1997; Armenia-Cope et al., 1993). Based on previous single and combined axis studies NHTSA researchers postulated that the two most critical modes of neck injury were axial force (tension or compression) and fore/aft bending (flexion or extension) (Cheng et al., 1982; FMVSS-208 200; Mertz and Patrick, 1971; Nusholtz et al., 2003; Eppinger et al., 1999). Associated critical values for each load (Fz and My) are used in the calculation of the Nij and are different based upon subject body mass. An Nij value below 1 limits risk of neck injury to a 22% risk of AIS 3 injury (Eppinger et al., 2000) As seen in Equation 2, the Fint and Mint values are the critical
values established for the maximum tension (+Fz) / compression (-Fz) and flexion (+My) / extension (-My) (Eppinger et al., 1999; Mertz et al., 1978)
30
Equation 2. NHTSA Nij Neck injury Criteria
The inadequacies of the Nij in escape system testing have been identified by Parr et al. (2013). New neck injury criterion developed by Parr et al., which will be called the MANIC(Gx) in this thesis to distinguish it from the Nij, has been compared to the Nij (Parr et al., 2013). The evaluation and comparison of Nij risk curves to MANIC (Gx) curves indicated that the Nij risk curves are inadequate for use in the aviation domain. The AFLCMC escape system oversight office has stated the desire for neck injury criteria that limit injury to a 5% risk of AIS 2 or greater (Parr et al.,2013). This is more
conservative than the Nij limit of 1.0 used by NHTSA, which equates to a 22% risk of AIS 3 neck injury (Eppinger et al., 2000). A pilot may have to flee from capture upon landing while passenger vehicle accidents typically receive attention from emergency response crews, for this reason it is prudent to assign a more conservative injury criterion to the aircraft ejection environment.
BEAM Criterion
The Beam Criterion (BC) was developed by Bass et al, and intended to be used in -Gx impact crash tests, specifically accounting for helmet mounted mass (Bass et al., 2006). The researchers used 36 head and neck complexes (portions of a full body cadaver) along with six whole cadavers. The head mounted mass was varied as well as
31
the CG location. The cadavers were tested along with a set of matched THOR and Hybrid III ATDs. Injury and non-injury data points were pulled from these matched tests to produce risk functions and assign a BC value to a 50% injury intercept through
survival analysis. Initially due to lack of BC critical values, the Nij 50th percentile Hybrid III ATD critical values were used. After risk function generation a second
iteration was used to minimize the standard deviation of the 50% intercept and normalize the BC to a value of 1. To achieve the BC of 1, Fz and My value ratios were altered, shifting these ratios resulted in values that compared well with previous NHTSA
intercept values. These risk functions ultimately resulted in the Beam Criterion creation. The equation to determine the BC is similar to the Nij in structure and it initially attempted to account for multi-axis forces that are present in the C7/T1 intervertebral disc at the lower neck, including shear forces (Bass et al., 2006). The shear was determined to be a non-factor in this test scenario so it was removed. The researchers also postulated that the Nij criterion was not suitable for Hybrid III ATDs while supporting head supported weight, stating that the Hybrid III was not designed for testing under these circumstances. Bass concluded that THOR ATDs responded more similarly with the PMHS during sled tests. The upper neck flexion moment was also found to be lower than the lower neck response, due to the moment arm being exacerbated by additional helmet weight at the C7/T1 intersection.
32
NIC
Navy and Air Force researchers developed a comprehensive injury evaluation tool that incorporates upper and lower neck loading by combining 12 separate neck injury criteria (NIC) that were used in the aircraft escape community (Nichols, 2006). The NIC is intended to be applied to current and future HMD escape systems focusing on
preventing injury. Nichols stated the inclusion of upper and lower neck loads was necessary as injuries have been known to occur at both locations during ejections. The NIC is comprised of six lower neck and six similar upper neck evaluation measures, they are: tension duration (+Fz), compression duration (-Fz), resultant shear duration (Fx, Fy), Nij (explained previously), maximum instantaneous lateral bending (Mx), and maximum
instantaneous twisting (Mz). These six criteria provide data and limits for all 6 degrees of freedom in both the upper and lower neck response but they have been aggregated to be used as a “success criteria” not as an ejection safety pass fail indicator (Nichols, 2006). Instead a failure (exceedance of the limit) in one of the six areas listed previously is considered an indicator “flag” and that exceedance requires further evaluation by a subject matter expert (SME). Because of these facts, there is ambiguity built into the NIC criteria and its
interpretation making it difficult to use as a system performance measure in developmental testing (Parr, 2014). There is a possibility of passing and failing a test in the same degree of freedom due to the way the duration and instantaneous load values are reported. A more in depth explanation, has been provided elsewhere (Parr, 2014). Multiple researchers have suggested that this tool is not valid for escape system level evaluation because of the
inconsistencies stated here (Carter et al., 2000; Pellettiere et al., 2011; Pellettiere, 2012; Parr et al., 2014). This criteria has been shown to be most effective at predicting lower neck
33
injuries in PMHS tests while the Nij under-predicted and the BC over-predicted observed injury during tensile force testing (Salzar et al., 2009).
Tensile Neck Injury Criterion
Neck injury criteria were originally developed for automotive frontal impact testing that indicated the most common mode of injury was frontal flexion. Tension not flexion is the primary loading mechanism in aviation and Carter investigated the development of alternate evaluation criteria (Carter et al., 2000). The two phases of ejection where tensile forces are most significant are during wind blast and parachute opening shock (Parr, 2014). This study used a combined human/PMHS data set to develop risk curves using logistic regression that were body size specific. These risks curves show that a tensile load of 2320 N for a large subject and 1740 N for a small subject indicate a 5% probability of AIS level 3+, this finding is validated by other
studies (Carter et al., 2000). Carter et al. highlight the limitation of their single-axis injury criterion when applied to ejection environments with multi-axial input. Parr et al. (2013) provided an update to the Carter et al. tensile criterion by including additional PMHS data into the analysis and generating risk functions using survival analysis. The researchers’ future research recommendations focus on incorporating multi-axial
components for improvement of the Tensile Neck Injury Criteria. The following section summarizes these improvement efforts and culminates in the Multi-Axial Neck Injury Criteria (MANIC) intended for use as the USAF testing standard.
34
MANIC
The MANIC is an improved neck injury criteria developed by Parr (2014) based upon previous research performed by Perry and others (Perry et al., 1997). The previous research highlighted the need for development of multi-axial, pilot-scale neck injury criteria that could be used to assess new and legacy escape systems with incorporated HMDs. In an effort to meet the USAF escape system oversight office’s requirement for neck injury, a multi-axis neck injury criterion, the MANIC, was developed. The initial formulation of this criterion is shown in Equation 3.
Equation 3. Multi-Axial Neck Injury Criterion (MANIC) (Parr, 2014)
Force components and bending moment components are compiled in a root sum of squares format using force and moment data that has been recorded during single axis tests to create risk curves and evaluate for injury in each specific input axis; resulting in a MANIC(Gx), MANIC(Gy) , and a MANIC(Gz). These combined values can be looked at as a single evaluation criterion, and based upon developmental test results a pass or fail determination can be made since the risk functions will provide injury risk and
classification information. MANIC data components were missing in some of the three acceleration planes due to a lack of data in the literature. The logic behind these
35
creation of this criteria refer to the original research (Parr, 2014). The resultant Gx, Gy, and Gz equations are shown below in Table 1 along with the applicable limits.
Table 1. MANIC Components (Gx, Gy, Gz) and limits (Parr, 2014)
Ultimately, the missing data did not affect the ability for risk curve creation, which is the core of neck injury risk prediction and the most usable tool for decision makers. The MANIC injury criteria can be used to generate a set of three human-based risk functions in each input acceleration plane Gx, Gy and Gz. Specific AIS level risk functions can be created by using the recorded human and PMHS neck response, for MANIC research AIS level 2 or greater (AIS 2+) and 3 or greater (3+) risk curves were created for comparison and evaluation. A 5% injury probability limit can be applied when using the MANIC risk curves, instead of the 22% for the Nij (Parr, et al., 2013). Each of the three specific input conditions produces varying amounts of head and neck movement during testing. These neck movements are the result of kinematic response, bracing, muscular contraction (human only), test setup, and subject positioning. The first application of this criterion was a test case completed by Parr in 2013 in the -Gx
36
acceleration axis, comparing the differences between NHTSA and MANIC(Gx) human risk curves, some of these comparisons are shown in Figure 4, Figure 5, and Figure 6.
The MANIC risk curves use survival analysis (SA) methods that have been applied to the field of injury biomechanics environments in previous research (Hosmer et al., 2008; Cutcliffe et al.,2012; Bass et al., 2006). SA uses human sub-injurious data combined with an injurious PMHS data set creating a dichotomous data set for SA evaluation. This method requires noting the AIS level associated with the injury points to form a risk curve similar to the one shown in Figure 6. Parr applied human study data as well as PMHS and ATD data to each axis of acceleration to develop three separate risk curves (Parr, 2014). With the human subjects experiencing non-injurious 6-10 G accelerations and the PMHS and ATD subjects experiencing 8-45 G’s, four of these resulting in injury (four of the six PMHS). The point of injury for the PMHS were not known until post test examination, zero human injury (AIS2+) were noted, creating a dual (right and left) censored data set. SA is capable of accounting for left and right censored data and is more applicable in this domain than logistic regression methods which made it the appropriate tool for Parr’s research.
The MANIC provides a means to evaluate the neck response during all four ejection phases. It defines injury risk using specific experimental force input data, or a combination of forces, resulting in a definitive injury prediction (Parr, 2014).
37
Table 2. MANIC Risk Function Input Matrix (Parr, 2014)
The use of the complete MANIC criterion will not be applied to this research but the Gx portion (MANIC(Gx)) shown in Table 2 will be utilized, concentrating on ATD to human neck response relationships. Based upon the aforementioned literature; aviation specific MANIC(Gx) limits that have been identified (<5% P(AIS 2)) and will be utilized for the remainder of the research. The methods and implementations of these approaches will be discussed in Chapters III and IV.
The figures below show a graphical depiction of the differences in injury probability between MANIC(Gx) and Nij (Figure 4 , 5 and 6). Similar construction characteristics and the common use of tensile and flexion limits allow for direct comparison between the two.
38
Figure 4. Probability of AIS 2 or Greater NHTSA and Human Neck Injury Risk Curves (Parr et al., 2013)
Figure 5. Probability of AIS 3 or Greater NHTSA and Human Neck Injury Risk (Parr et al., 2013)
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 1 2 3 4 5 6 P( ≥AIS 2) Neck Injur y MANIC(Gx)/ Nij NHTSA Risk Curve Human Risk Curve 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 1 2 3 4 5 6 P( ≥AIS 3) Neck Injur y MANIC(Gx)/ Nij Human Risk Curve NHTSA Risk Curve
39
Figure 6. Human AIS 2+ and AIS 3+ Risk Curves (Parr et al., 2013)
Statistical Methods
Statistical evaluation is a very important component of any research that is focused on expanding the existent research base. The evaluation of neck injury during ejection requires sound quantitative statistics, ensuring that the results will hold up under scrutiny. Linear regression is a practice that is most commonly applied to how two variables relate by fitting a linear equation to observed data. Human and ATD linear models will be useful in this research. In Chapter III and IV it will be used to characterize a human to ATD transfer function method.
Summary
In summary, no single neck injury criterion has emerged for use as an injury predictor during ejection, with or without head supported mass. Human kinematics have
0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 5 6 Pr obability of Neck Injur y (%) MANIC(Gx) AIS 2 Human Risk Curve AIS 3 Human Risk Curve
40
only been observed and documented at low levels, this data is a limiting factor when attempting to predict injury in a highly dynamic high-G environment. Neck
biomechanics are complicated and have myriad of input factors potentially affecting human neck response during ejection. ATDs and PMHS provide a surrogate to evaluate, however, the internal reactions of human musculature, tendons and joints are difficult to emulate via ATD and PMHS experiments. Human testing with the same subject and input conditions have been known to be highly variable; removing this variability with an ATD is a task that lies outside of this research scope. Continued research in ejection kinematics is required in order to build a more accurate surrogate for ejection testing (Salzar et al., 2009).
Some of the areas that have not been agreed upon are related to applying the data, ATD bio-fidelity, testing equipment and methodology misrepresentations. Many
research entities have postulated dissimilar injury evaluation criteria. Without a common language, it has proven difficult for future researchers to communicate, which has complicated the verification of existing results. Speaking in the same vernacular will allow the transfer of ideas between industries and stakeholders, as well as reducing the amount of rework associated with confirmation of previously defined concepts. The future efforts to integrate these results and clearly define a single method of predicting neck injury associated with ejections are the most challenging aspect of this field of study. Evaluating existing data and developing an ATD-to-human transfer function is at the heart of this research and will be expounded upon in the next few chapters.
41
III. Comparison of Human and ATD Neck Response to Frontal Impact (-Gx) Acceleration
Chapter Overview
The following paper entitled
“
Comparison of Human and ATD Neck Response to Frontal Impact (-Gx) Acceleration” has been accepted for presentation at the 2016Industrial and Systems Engineering Research Conference. It describes the creation of a proposed transfer function between ATD and human neck response through the use of linear regression modeling. A short example application is also included. The paper is presented as it was submitted without removing or editing content.
Abstract
Helmet Mounted Display (HMD) technology has become commonplace in many military applications. The current research proposes a method to create a transfer function to enable the direct application of a previously developed human-based neck injury metric called the MANIC(Gx) to accelerative tests performed with anthropometric test device (ATDs) during prototype and developmental testing in the systems engineering process [1]. Data was collected from previous accelerative sled experiments where ATDs and human/post mortem human subjects (PMHS) were accelerated over a range of frontal (-Gx) loading and peak MANIC(Gx) was calculated. Linear regression was performed for each data set and the difference between human and ATD expected value for peak
MANIC(Gx) was analyzed. It was observed that ATD MANIC(Gx) expected values were lower than the human values at each level of input acceleration above 2 G. A human to
42
ATD transfer function method is proposed based on these observed differences. This transfer function could make the previously developed human-centric neck injury criterion directly applicable to developmental safety testing with ATDs [1].
Introduction
The proliferation of helmet-mounted displays has led to increased head supported mass in military fixed wing and rotary wing aircraft systems. This increased head
supported mass has in-turn increased the risk of neck injury. More specifically, the need for pilots to support additional weight on their heads can lead to acute neck injuries from instantaneous accelerative forces that occur during ejections or other crash-related events. With the ubiquitous integration of HMDs and other head mounted equipment (e.g. night vision devices and helmet mounted cuing systems) into the military operational
environment, a significant amount of consideration has been given to aviation specific neck injury criteria. However, current methods do not permit design engineers to accurately predict how new HMDs will affect the neck of the user in the dynamic and unpredictable aviation environment.
A United States Air Force (USAF) policy was previously issued that expands the acceptable size and weight range of future aircrew, causing special concern for the safety of aircrew at the smaller end of the spectrum. The additional injury likelihood for pilots weighing less than 136 lbs. has forced the F-35 Joint Program Office to restrict the use of this platform to pilots weighing at least 136 lbs. (USAF Public Affairs Office, 2015).
Integration of new technology often poses problems for end users if not
43
when a new piece of hardware is available for operational use. In a business environment, the added value and or utility are easy to discern; the costs of integration, training and hardware are always under evaluation, providing shareholder value is the primary concern. In these environments, a business case approach is commonly taken to make a cogent decision about the benefits gained (or lost) by integrating the new hardware, and a decision is made to invest in the new capability or to focus resources elsewhere.
In the Department of Defense (DoD), where profit is not a measurable concern, and where numerous stakeholders influence investment priorities, many challenges are present when facing advanced technology integration decisions. A Decision Maker (DM) in the DoD must be vet the decision with respect to likely future needs, considering both short term acquisition costs as well as longer term operations, maintenance, and disposal costs.
Integration of Helmet Mounted Display (HMD) technology is considered a crucial part of the future landscape of operations inside (and outside) the DoD [3]. Whether ground-based, or in the air, the user is forced to accommodate the additional head supported weight of any HMD hardware, while performing their mission. The benefit of potential operational advantages should offset the costs of potentially altered duty
performance and increased safety risk from supporting the additional weight. The current research focuses particularly on the increased risk of injury due to increases in head supported weight, although it is recognized that chronic stress injuries and fatigue are also of concern. Specifically, this research seeks to create a transfer function between human and ATD response to -Gx accelerative input, enabling the direct application of a
44
previously developed human neck injury criterion (with a comparable structure to the Nij) to accelerative tests performed with ATDs to ultimately steer future helmets towards a safer, more functional design [1]. Development of HMDs must account for the
downstream effect that the increased hardware and resulting increased head supported mass will have on the user. If these effects of new hardware hinder performance and add increase neck risk during ejection, a negative cost to benefit ratio is created. The cost to benefit ratio of new technology must be thoroughly scrutinized before these systems are integrated into legacy systems or incorporated into new development efforts. This scrutiny must consider the likelihood of human injury as an important aspect when quantifying costs and operational advantage as benefits.
To capture the risk, first the relationship of ATD and human biomechanics must be quantified. This study focuses on the ATD to Human relationship in the -Gx direction only. Multiple cadaver and ATD automotive crash tests have shown that the critical forces present in a neck during -Gx loadings are tension (+Fz) and bending moment (My) [4, 5]. These findings have propelled the aviation community to develop an aviation specific neck injury criteria that can be utilized to translate the neck responses observed in ATDs to humans. This neck injury criterion was developed by Parr et al. and has a similar structure to the Nij formulation but has different risk functions than the Nij, thus it is a distinctly different neck injury criterion from the Nij [1]. For this reason, hereafter it will be referred to as MANIC(Gx), where MANIC stands for multi-axial neck injury criterion.
45
This paper seeks to answer the following research question: what is the
difference in peak MANIC(Gx) between human/PMHS and ATDs over the range of -Gx accelerative input observed from previous laboratory experiments? An additional question to be answered is: can the observed differences in peak MANIC(Gx) be used as a transfer function to make the Parr et al. human-based neck injury criterion more appropriate for use in testing with ATDs?
Background
The Air Force Life Cycle Management Center (AFLCMC) has requested the development of a multi-axial criterion for evaluating pilot safety during ejection. The current evaluation techniques do not provide a stand-alone solution for this problem. Developing the state of the art neck injury criteria has proven to be an iterative process. This new evaluation criteria is to be used in future injury prediction calculations at the program level during prototyping and developmental testing [1].
The automotive industry created the basis of the neck injury criterion that has been used in aviation applications. The National Highway Traffic Safety Administrations (NHTSA) has been studying the injurious conditions that are present in frontal vehicular impacts for decades; and have a vast amount of data and test results available for review and application to ejection. There are similar modes of neck injury prevalent in
automotive accidents and aircraft ejection [6, 7]. Decomposing the phases of ejection into singular force vectors allows a comparison between ejection modes of injury and automotive crash tests. Ejection has distinct phases: catapult, windblast, drogue-chute deployment, main-chute deployment and seat-man separation,, and free fall.
46
Windblast, droguechute deployment, and mainchute deployment can create a -Gx acceleration on a pilot that is very similar to crashing an automobile [1]. NHTSA developed the Nij in an effort to accurately assess the human neck response during a frontal impact to evaluate new car airbag safety [4]. After multiple test were completed and examined, it was determined that human neck injury was most likely to occur as a result of +Fz (tension) and +My (flexion) forces which occurred during such a frontal collision [4]. A specific formula and intercept values (Fint and Mint) were also developed
to provide a single metric, referred to as Nij, which could be applied across multiple
passenger weight ranges [4]. For complete details on the Nij, the reader is referred to publications by Eppinger et al. [4, 5]. However, the Nij formulation is shown in Equation 2. NHTSA Nij Neck injury Criteria
(1)
However, this formula alone does not provide information regarding the likelihood or severity of human injury during an impact. Injury severity is captured through the use of the Automated Injury Scale (AIS). This scale provides researchers a consistent method for classifying each injury according to its severity on a 6-point ordinal scale, with the least severe injury labeled AIS 1 and the most severe labeled AIS 6 [8]. The AIS provides a means for scoping analysis effort as well as a precise way to present injury data to decision makers once it has been analyzed. Limiting ejection injuries to below AIS 2 or AIS 3 levels is common in USAF risk analysis; therefore these levels are commonly used during neck injury risk analysis. Years of extensive NHTSA research with the Nij has resulted in a set of risk functions generated with logistic regression that
47
predict probability of injury (AIS 2 through 6) based upon an observed Nij value (Eppinger et al., 1999; Eppinger et al., 2000).
The USAF has previously used a notional guide for design, development, and acquisition of HMDs called the Knox Box. The Knox Box, developed by USAF ejection researchers was an effort to provide guidance to US Air Force Research Laboratory and USAF acquisition program personnel who were developing HMD prototype designs [9]. The Knox Box uses mass and CG as its main metrics for design constraints, creating a virtual box that all HMD prototype designs should fall within. The goal was to limit the weight and off-axis moment (CG) to minimize chances of pilot neck injury during catapult phase of ejection. This criterion was developed as an early design guideline, not intended for long-term use in HMD designs however; it is still being utilized as the starting point for HMD design [9].
The US Navy and Air Force have developed a 12-part Neck Injury Criterion (NIC) based on automotive injury criteria. The NIC uses instrumented aerospace ATDs of varying sizes configured with a Hybrid III neck. Those ATDs were subjected to sled test ejections at differing equivalent airspeeds to permit the quantification of likely head accelerations and neck forces which a pilot might encounter during an ejection while wearing a prescribed helmet and when ejecting with a specific ejection system design [6]. The NIC is currently the requirements generation tool of choice in HMD and ejection seat development for DoD fixed wing aircraft. Although the NIC is currently being used by the F-35 program for ejection evaluation testing, past research has highlighted some limitations of the NIC [1]. One of the most significant concerns is inconsistent limits for
48
the various neck loads included in the 12-part criteria. This criteria includes some elements that are not linked to specific AIS injury levels nor supported by underlying injury risk curves to allow quantification of the risk percentages associated with a specific level of observed loading [10]. The use of the Hybrid III neck in these evaluations is noteworthy as this mechanical structure has been designed to respond similar to the human neck during accelerative events, however, some research has indicated that its response is not always consistent with the response of the human neck (ref).
Attempts have been made to account for the discrepancies in the above methods and to fulfill the requirements set forth by AFLCMC. One proposed method, the Multi-Axial Neck Injury Criteria (MANIC), was designed to include as many of the six major forces that could be observed in the upper neck as a result of accelerative loading as necessary [1]. The MANIC is the only proposed neck injury evaluation tool that fulfills all of the requirements that have been specified by the Air Force Life Cycle Management Center for a neck injury criteria. However, the MANIC provides human risk curves for each primary axis of acceleration (Gx, Gy, and Gz) constructed using a combination of data from accelerative tests with human and post-mortem human subjects (PMHS) and application of these human risk functions can only be applied during ejection tests using ATDs if it is assumed the ATD and the mechanical neck in the ATD respond equivalent to a human pilot, an assumption the literature has already demonstrated is flawed [1,6,7] Therefore this research will seek to create a modified MANIC(Gx) metric, which is
